Monoamine oxidase (MAO) activity and the average evoked response measure of augmenting/reducing (AR) differentiate some psychiatric patient groups from normal controls. This study tested whether a combination of MAO and AR predict psychiatric vulnerability in a nonhospitalized and previously undiagnosed sample. Almost 400 college males were screened on the Minnesota Multiphasic Personality Inventory (MMPI). A subsample of students with indications of psychological problems was selected as was a control sample. These students were classified into 4 biologically homogeneous groups--low MAO/augmenting, low MAO/reducing; high MAO/ augmenting, high MAO/reducing. Based on the Research Diagnostic Criteria (RDC), more psychopathology is found among low MAO/augmenting and high MAO/reducing students. This result replicates the pattern indicated in a previous study and suggests that specific biological variables may be related to a general vulnerability to psychiatric problems.